Gender and Spatial Heterogeneity in the Impacts of Covid-19 on Households’ Income in Ethiopia: Evidence from High Frequency Phone Survey

This paper examines the existence or otherwise of gender and spatial heterogeneity in the impacts of the early days of the Covid-19 pandemic on households’ total incomes in Ethiopia. Using the first round of the World Bank’s high frequency phone survey on 3224 households, the logit model regression analysis finds that households in different regions of the country are affected disproportionately and heterogeneously. In order to examine the existence of heterogeneity, a logit model containing interaction terms between gender, and region dummies six Covid-19 containment measures have been fitted on binary outcome of change in total income of households. Linktest of specification, Hosmer-Lemeshow goodness-of-fit test, VIF and tolerance index of multicollinearity tests and various influential observations tests are conducted in order to fit models that can handle the heterogeneity analysis of the impacts of the pandemic on households’ total income in Ethiopia. The results suggest the existence of statistically significant regional heterogeneity in the impacts of within country travel restriction, limit on social gatherings and closure of schools and universities on the total income of households. Restriction on international travel, curfew/lockdown and closure of non-essential businesses do not bring heterogeneous impacts across gender and region of households. Gender of the household head is insignificant determinant of change in households’ income and also does not cause heterogeneous impacts. Overall, in Ethiopia there exists region-based heterogeneity in the impacts of Covid-19 on households’ total income. The results imply the relevance of income policy measures that can lessen shocks to household’s income, and livelihood after a pandemic such as Covid-19.